Abstract

Total Maximum Daily Loads (TMDLs) have been established under authority of the Federal Clean Water Act for the Snake River-Hells Canyon reach, on the border of Idaho and Oregon, to improve water quality and preserve beneficial uses such as public consumption, recreation, and aquatic habitat. The TMDL sets targets for seasonal average and annual maximum concentrations of chlorophyll-a at 14 and 30 micrograms per liter, respectively. To attain these conditions, the maximum total phosphorus concentration at the mouth of the Boise River in Idaho, a tributary to the Snake River, has been set at 0.07 milligrams per liter. However, interactions among chlorophyll-a, nutrients, and other key water-quality parameters that may affect beneficial uses in the Snake and Boise Rivers are unknown. In addition, contributions of nutrients and chlorophyll-a loads from the Boise River to the Snake River have not been fully characterized.

To evaluate seasonal trends and relations among nutrients and other water-quality parameters in the Boise and Snake Rivers, a comprehensive monitoring program was conducted near their confluence in water years (WY) 2009 and 2010. The study also provided information on the relative contribution of nutrient and sediment loads from the Boise River to the Snake River, which has an effect on water-quality conditions in downstream reservoirs. State and site-specific water-quality standards, in addition to those that relate to the Snake River-Hells Canyon TMDL, have been established to protect beneficial uses in both rivers. Measured water‑quality conditions in WY2009 and WY2010 exceeded these targets at one or more sites for the following constituents: water temperature, total phosphorus concentrations, total phosphorus loads, dissolved oxygen concentration, pH, and chlorophyll-a concentrations (WY2009 only). All measured total phosphorus concentrations in the Boise River near Parma exceeded the seasonal target of 0.07 milligram per liter. Data collected during the study show seasonal differences in all measured parameters. In particular, surprisingly high concentrations of chlorophyll-a were measured at all three main study sites in winter and early spring, likely due to changes in algal populations. Discharge conditions and dissolved orthophosphorus concentrations are key drivers for chlorophyll-a on a seasonal and annual basis on the Snake River. Discharge conditions and upstream periphyton growth are most likely the key drivers for chlorophyll-a in the Boise River. Phytoplankton growth is not limited or driven by nutrient availability in the Boise River. Lower discharges and minimal substrate disturbance in WY2010 in comparison with WY2009 may have caused prolonged and increased periphyton and macrophyte growth and a reduced amount of sloughed algae in suspension in the summer of WY2010.

Chlorophyll-a measured in samples commonly is used as an indicator of sestonic algae biomass, but chlorophyll-a concentrations and fluorescence may not be the most appropriate surrogates for algae growth, eutrophication, and associated effects on beneficial uses. Assessment of the effects of algae growth on beneficial uses should evaluate not only sestonic algae, but also benthic algae and macrophytes. Alternatively, continuous monitoring of dissolved oxygen detects the influence of aquatic plant respiration for all types of algae and macrophytes and is likely a more direct measure of effects on beneficial uses such as aquatic habitat.

Most measured water-quality parameters in the Snake River were statistically different upstream and downstream of the confluence with the Boise River. Higher concentrations and loads were measured at the downstream site (Snake River at Nyssa) than the upstream site (Snake River near Adrian) for total phosphorus, dissolved orthophosphorus, total nitrogen, dissolved nitrite and nitrate, suspended sediment, and turbidity. Higher dissolved oxygen concentrations and pH were measured at the upstream site (Snake River near Adrian) than the downstream site (Snake River at Nyssa). Contributions from the Boise River measured at Parma do not constitute all of the increase in nutrient and sediment loads in the Snake River between the upstream and downstream sites.

Surrogate models were developed using a combination of continuously monitored variables to estimate concentrations of nutrients and suspended sediment when samples were not possible. The surrogate models explained from 66 to 95 percent of the variability in nutrient and suspended sediment concentrations, depending on the site and model. Although the surrogate models could not always represent event‑based changes in modeled parameters, they generally were successful in representing seasonal and annual patterns. Over a longer period, the surrogate models could be a useful tool for measuring compliance with state and site-specific water‑quality standards and TMDL targets, for representing daily and seasonal variability in constituents, and for assessing effects of phosphorus reduction measures within the watershed.

First posted December 15, 2011